Can 3-d printing be called scratch-building?

[Kris]

So enlighten me. What is 3D printing In laymans terms i.e. I want a print of a Dock Tank bodyshell say. How can I do that? What is the result printed on? Not paper surely as paper is 2D. Am I missing something?

 

The printer lays down a thin layer of plastic / resin / metal*. Once this layer has been put down it goes back to the start and puts down a second layer on top of the first, then a 3rd, 4th, 5th etc. These build up to create a 3d shape. The printer gets the information from a 3d CAD drawing, so if you want your dock tank body shell you need to find the 3d CAD drawing or create this yourself. The design is not a quick process and printing is slow. It's still very expensive and the resolution is not fantastic at the affordable(ish) end of the market.

 

*dependant on printer

 

This link gives more information.

 

 

 

[halfwit]

I went to the model engineering exhibition at Sandown Park yesterday, and notices a sign by a very nice aero-ebngine, to the effect of 'no CNC machining used in this model' which struck me as an interesting parallel. CNC clearly being seen as cheating!

 

Jon

 

Interesting view. I used to program CNC lathes, as well as turning manually. With CNC you still need to set the tooling up, select the relevent speeds and feeds, set the job in the chuck and write the program which will produce the componant without anything going bang (tooling hitting the chuck/workpiece, never happened to me but I did see some spectacular crashes). So its still a highly skilled process, just a different way of achieving the end result. Obviously there were some jobs that could only be done manually (at least where I worked.)

[Guest jim s-w]

This tool when it has fully evolved, please note the emphasis, will enable many people to create models they cannot currently dream of creating. Either they will acquire the skill to use it directly themselves or they will pay for someone to do it for them, because it will be relatively cheap. Compared, that is, to the current alternative of having a highly skilled person create it for you.

 

Thing is the highly skilled person is still in the chain. Whether or not you learn to do it yourself or pay someone this technology changes nothing. It certainly wont be any cheaper. Think of it like this - swap the CNC/stereolythography stuff for etching. I have researched many differnt manholes for my layout. I had the software and the skills to draw this up as artwork for an etch, If people want a set i can supply one but you will have to pay for it (I dont do this as a business so what you pay is pretty much cost). What if you want the artwork so you can etch your own? IF i decide to sell it you, bearing in mind that you can run off as many as you want and sell them to your friends I would charge you a hell of a lot more for the artwork - Even then you wouldnt own the phototool and would have to get your own made plus have sheets etched. You have suddenly gone from paying a tenner for a set to paying a couple of hundred pounds for the same thing.

 

Its kinda like custom weathering. the true pros - the ones with a very high quality output - will cost you a fortune, there will always be people who will undercut them but you will get what you pay for.

 

Lets say you want a simple tank wagon.

 

Can you find a prototype?

Can you get permission to access it?

Do you have the skills to accurately survey it?

what about the skills to transfer that from your measurements to a 3D cad rendering?

 

This technology only changes the last step, people will still need an equal amount of skill to produce model.

 

Cheers

 

Jim

[Guest jim s-w]

Although some may question whether getting etchings produced was scratchbuilding, the attitude you have shown, of just getting on with things and solving problems as they arise, is identical - it is far too easy to get sidetracked into foreseeing and "solving" every problem and hence to have made no concrete progress, as I know only too well from my own navel gazing.

 

Thanks, ultimately I am not that fussed how I get the end result. As I have said before if Scaledale decided to produce a model of the rotunda I would probably just use that!

 

Cheers

 

Jim

[Lady_Ava_Hay]

The printer lays down a thin layer of plastic / resin / metal*. Once this layer has been put down it goes back to the start and puts down a second layer on top of the first, then a 3rd, 4th, 5th etc. These build up to create a 3d shape. The printer gets the information from a 3d CAD drawing, so if you want your dock tank body shell you need to find the 3d CAD drawing or create this yourself. The design is not a quick process and printing is slow. It's still very expensive and the resolution is not fantastic at the affordable(ish) end of the market.

 

*dependant on printer

 

This link gives more information.

 

 

 

 

Fascinating link and info. I think the powder and inkjet technology is the most likely to succeed but, assuming the costings to be in the £10's of thousands and the CAD CAM being as time consuming as it is, I cannot see it being a good living for anyone.

 

Am I right in thinking that Dapol used some imaging style of X Ray on a Thumper to produce the CAD Cam for that?

 

I can see that the first to produce a desktop 3D printer for less than a £1000 might well have a lot of takers and one could produce a rake of mineral wagons in an afternoon using this.

[t-b-g]

Not directly related to 3D printing but to the other "new" technology of laser cutting of parts. A good friend of mine showed his wife a beautiful laser cut building that had been commisioned by somebody who had done the drawings and sent them off to have them made.

 

Her reply (very deeply thought provoking for a non model enthusiast!) really opened my eyes. "It is very good but it is not model making".

 

I don't know if model railways needs a definition of scratchbuilding apart from those times when an entry of a model into a competition requires one. As has already been said, true scratchbuilding should need everything to be hewn from raw materials and that was always very rare.

 

My view is that if a model can be duplicated by having more manufactured by a third party, then scratchbuilt it ain't. I cannot see any logic in the arguement that if a kit designer builds his/her kit it is scratchbuilding. Why isn't it them building their own kit!

 

On the other hand!!!!!! I recently built a loco and made some parts by creating plasticard masters and then a friend helped me to profile mill them. Another friend built a loco using some parts from various kits and a kit tender to create a loco that is not available as a kit, only hand making quite a small number of bits. Scratchbuilt or kit conversion? I don't know and I don't really believe that it matters one jot! How about we call them "handmade".

 

To me the joy of modelling is picking up tools and materials (etched, cast or raw sheet doesn't matter a bit) and using my small amount of skill to turn them into something that looks like it is supposed to! Drawing something on a computer and having somebody else print it in 3D is very clever use of available technology but is it model making? To me it is designing a component and commissioning a manufacturer to produce it. Having said that, it is true that each of us has the total right to enjoy our hobby as we see fit and just because it is how I see it, I would never expect more than a few to share my views. Tin hat on! Sits back and waits for missiles.............

[tebee]

With a decent Cad package it not that horrendous a job to draw things in 3-d as it was a few years ago. To give an example I started out wanting to models of the Isle of Man 4 wheeled coaches for my freelance English 3 foot gauge line. Now I have Boyd's book with drawings of the A,C and E class versions of these but to make life difficult I wanted A,B and the MNR version of the E class. I also had a friend who wanted some Irish Bogie coaches but we had only overall dimensions and photographs of these.

 

Now one of the nice things about cad is that you can cut and past so you only need to draw something once. With the package I used it's possible to to take 3-d parts and combine them in to other parts. So the approach I took was to divide the coach side into a number of slices, draw each one of them, then combine a number of them to make the final side. It's a bit like going back to the old days and doing cut and shut on Triang clerestories, but virtually! It's also possible derive a component from an existing component so you can take a coach side and derive a mirror imaged version of it for the other side or you can take a 15 inch long piece of paneling and derive a 9 inch long one from it.

 

So by using these short cuts and working 4-6 hours per day it took me roughly 3 days to do the sides for my three IOMR coaches but included in that was learning the package itself as I'd not used it before . Roofs and ends took about another day but buy using the pre-existing components the 3 Irish coaches took only 1 day - and I also did another 3 freelance variants in that time.

 

Now I've made some compromises, The Irish coaches are to a certain extent guesstimates and I've ignored the differences in the paneling style between the two, but the thing is, if someone wants an completely accurate version of one of those coaches and can supply me with the relevant information I only need to change my drawings and the next one made will be correct at no other cost. There are no molds or masters to re-make.

 

This leads to an interesting possibility, would it be possible to make models the same way we make open source programs? People freely give their time to contribute to a community project where anyone with a few basic skills can add their improvements to the final product. Even those who have no skills can contribute just by explaining to the people who do what is wrong.

 

I also wonder if there will be a merging of those who model virtually - the people who make content for the train sims, and the mainstream modelling community.

[hollywoodfoundry]

I would have thought that scratch-building implies that the person doing the building produced all or most of the components of the model themselves. In practical terms, that may not always be possible, as you will most likely have to buy in the motor, gears and possibly the wheels, although around 50 or 60 years ago you would most likely have built those as well.

 

To draw up the artwork for an etch or produce the 3D artwork for either 3D printing or a stereolithographic build, the etching and builds are being made by someone else. Or perhaps a machine. So even though you may be the 'author' of that work, you did not build those components yourself. To claim the result is scratch-built is stretching the definition, in my opinion.

[The Q]

My randon thoughts on this.

Today people claim they built their own computer, by plugging a few boards in. But I remember soldering chips together to do it, (and I worked with someone who could and did design his own boards for his computer). But would I go back to building boards no...

 

On my railway, if it exists as a manufactured item I don't make my own but I will upgrade the RTR item. Even though I like scratch and Kit building.

 

If I decided to have a 3D printer it would take the cost/time / effort out of build a large number of Identical Items but I wouldn't describe them as scratch built, but more like a home made Kit.

 

See http://www.reprapcentral.com/ it appears 3D printers are getting well with reach of many of us.

 

It only needs time before there is a collection of CAD files on the net of railway items, donated by the software capable people here. A 3D printer and voice activated computer, Then it would be as easy as Captain Picard would say " Tea Earl Gray hot"

And no doubt some would say they made it themselves..

 

The Q

[hollywoodfoundry]

See http://www.reprapcentral.com/ it appears 3D printers are getting well with reach of many of us.

The Q

 

Unfortunately the RepRap produces truly awful results, I doubt you would be even varguely interested in what it produces. Even the best of the rapid prototyping techniques is only now producing results equal to moderately good resin casting. True, the cost will come down in time, but is is still cost-effective to have the builds done by a bureau.

 

Personally, I don't believe it is necessary to own such a machine, I use lot of gears, but I don't buy the gear hobbing machine, prefering to have someone else look after that part of the job.

[Rods_of_Revolution]

I'm a CAD artist and I have produced for myself almost 100 3D printed models, I also sell over 40 different models in my shop at Shapeways.

 

I would certainly say if you do the CAD work it is scratch building. The computer and the 3D printer are tools like a pencil and a craft knife. Saying it's not scratch building is like saying if you use a knife instead of your bare hands to cut a piece of plasticard it isn't scratch building. My hands tell the computer and printer what to do as my hands tell a knife what to do when I am cutting with it.

 

Computers, 3D printers, knifes, drills etc are all tools.

 

Kindest Regards,

 

Jack

[Neil]

.... I would certainly say if you do the CAD work it is scratch building. The computer and the 3D printer are tools like a pencil and a craft knife. Saying it's not scratch building is like saying if you use a knife instead of your bare hands to cut a piece of plasticard it isn't scratch building. My hands tell the computer and printer what to do as my hands tell a knife what to do when I am cutting with it. ....

 

I'm afraid I still remain to be convinced. I'll use a recent example to explain why.

 

Earlier this year I scratchbuilt the body for this narrow gauge loco. I mainly used knife and plasticard, though there are a few bits of brass and a commercial cast chimney in the mix.

 

 

This green loco is destined for my next 009 layout. At the same time I built another example in blue, to sell on.

 

 

Again the same recipie was followed, the majority of the body being constructed from sheet plasticard. Both locos employ Kato mechanisms.

 

Now if I'd used CAD to originate the drawing for 3-D printing; I'd have done the work for the first one, but the second and any subsequent versions would just have been an exercise in pushing a button. Effectively I'd be donig what the bods on the Bachmann injection moulding machine in China do.

 

There is another parallel that I'd also like to wheel out. If I didn't have the ability to make sttuff for myself, yet want something unique then assuming I had deep pockets I could commission a craftsman to make that thing for me. To ensure that I got exactly what I wanted then I'd need a detailed specification and drawing that I could give to the craftsman for him to work from. It strikes me that this is what is happening with 3-D printing, the part of the craftsman being taken by the machine.

 

I'm sure that this debate is fuelled by the sneaking suspicion that scratchbuilding is somehow the pinacle of our hobby and that other forms of endevour are somehow less worthy of our admiration. I'm certain that this shouldn't be so, I'd like to see skills valued for what they are not have to bask in some form of assumed reflected glory. I'm gobsmacked by some of the results of bespoke etching and casting that I've seen and once the surface finish issues of 3-D printing are resolved then I'm sure I'll be equally gobsmacked by them too.

 

Last thought (hurrah), the stone walling in the lower photo of the blue loco, is a plaster casting that I produced from my own moulds and masters. It's cast not scratchbuilt.

[Rods_of_Revolution]

Say I produced a model using a 3D print with added brass wire handrails, chassis, painted and finished. It's one of a kind and I designed the model, used a 3D Printer to produce the parts and painted/finished it, how would you describe it?

 

I personally think scratch built is the nearest term to describe it. It may be a kit of parts but so are the plasticard pieces once they are cut from the sheet.

 

Here is an example of a model I built as per the above description, a Macedonian Railways class 661.

 

 

 

Kindest Regards,

 

Jack

[Neil]

Nice model Jack; what you've done is not disimilar to the route I took to producing a brace of class 24's.

 

 

This one and its other two companions use modified Hornby bodyshells mated to modified Bachmann mechanisms. They too feature additional wire bits for handrails and pipes along with some hand cut flush glazing. However they're not scratchbuilt, they're a hack of readily available bits into something more 24 like than the rtr offering. Your Macedonian diesel appears to follow identical principles apart from some clever drawing stuff right at the begining which means that instead of sending off for a few bodyshells from East Kent Models (like me) you've pushed a button that's requested the 3-D printer to make you a bodyshell.

 

I'm not at all concerned that my models aren't scratchbuilt, it being how 24-ish they look that counts.

[RepRap Central]

Unfortunately the RepRap produces truly awful results, I doubt you would be even varguely interested in what it produces. Even the best of the rapid prototyping techniques is only now producing results equal to moderately good resin casting. True, the cost will come down in time, but is is still cost-effective to have the builds done by a bureau.

 

Personally, I don't believe it is necessary to own such a machine, I use lot of gears, but I don't buy the gear hobbing machine, prefering to have someone else look after that part of the job.

 

"Quality of parts currently achieved by a RepRap Printer"

 

http://blog.reprap.o...el-quality.html

 

Straight off the printer!!!

 

If you wish to look a little closer at this and other DIY/Home 3D Printers take a look at http://www.reprapcentral.com

 

From one Engineer to another, trust me, you are going to want to embrace this technology, its potential is incredible!

[Pugsley]

Nice, but still way off the mark for producing model parts, there's still too much visible texture from the layering process. For model part creation, I think you need a resolution of 0.1mm at most, even finer if possible. Some of the pro systems will do that, the Objet 24 being one that I know of, but these are still £12K, so out of most domestic users price range. I have no doubt that the home systems will reach that level eventually, but they've got a fair way to go.

 

I have my doubts that 0.1mm resolution can be achieved by extruding ABS, but I'm happy to be proved wrong in the future.

[RepRap Central]

I agree, there's a way to go to reach the level of quality you Guys need, but if you look at some of the 3D Printers we have, it won't be long before we're approaching that resolution and quality and at a tenth of the price of a commercial printer making them really affordable for the first time.

[hollywoodfoundry]

I agree, there's a way to go to reach the level of quality you Guys need, but if you look at some of the 3D Printers we have, it won't be long before we're approaching that resolution and quality and at a tenth of the price of a commercial printer making them really affordable for the first time.

 

That may be true if you move away from the extruding principle, otherwise I am afraid I can't see you reaching the resolution of the top quality commercial builds. I have several examples of items built using Fineline's Micro-resolution process, which they say is exclusive to them, and it is IMO way ahead of the others.

 

As I have said before, I don't need to own a printer, I have better things to do than drive one.

[RBE]

3D printing, scratchbuilding? I think not. That's not to denigrate the time, skills and effort put in at the keyboard, which quite frankly would be beyond me, but whatever it is, it isn't scratchbuilding. I understand scratchbuilding to be working in resistant materials, and while it's understood that some components can be bought in, and some materials can come in handy shapes and sizes (tube, angle etc) the bulk of the item starts out as something very different from the finished object. The best comparison I can come up with is the difference between a photo and a painting where both result in an image but the methods and skills brought to bear differ.

 

I don't agree, designing and drawing a 3D model of the thing you want to make and then having it printed is still creating your work using a tool. When 'scratchbuilding' in the traditional sense we still use extruded plastic, metal, different profile etc all of which have been designed by someone else and mass produced. The plastic used in these and in 3D printing all started off as something completely different. The model you are making is still coming from your research and your skill with a tool as much as cutting it from plastic profiles with a scalpel and gluing it together. As some have mentioned, if someone else then uses your 3D model and prints one off it then becomes commercial and not scratchbuilt for them. Much the same way as buying a Bachmann body for instance. But your models can IMO be classed as scratchbuilt as you made it. Virtually in the computer world initially and then physically as the finished model, but you still made it none the less. Your analogy of a photo compared to a painting is nonsense as a photograph is simply a captured image of a thing that already exists, there is skill in the composition etc but no creating of the object was invovled. There is significantly more physical input involved producing the painting of a similar object. (I am trying to be diplomatic to the photographers here as myself I am a keen photographer and painter so have a foot in each camp!). What you are effectively saying is because it is created on the computer first rather than physically carved from raw materials its somehow less valid. This is a major arguement that occurs in the art world regarding computer art vs traditional mediums. There is often a snobbishness from the traditionalists when confronted by computer generated art as if it is somehow cheating. This is frankly rubbish as the skills invoved are every bit as incredible whatever medium you use. Ok the computer has effects functions and buttons that do this and that but you still need the skill and creativity to generate a work of art, what the computer does is free up the artist to create his/her vision with a greater degree of clarity rather than getting bogged down with the limitations of the painting medium and techniques of physical application. What we really aspire to be as modellers are artists, sculpters, creators of a vision, a memory, a representation of life in something we can hold forever. In the pursuit of that memory we try to make our models as close to that image as we can. If we cant get them commercially we scratchbuild or convert. If that thing did not exist before we decided to make it then there is nothing to convert therefore its scratchbuilt, even if the 'kit' of parts is produced by a machine, as long as we designed the parts it holds true, whether we cut the shapes from plastic with a knife our tell the computer where to cut, build, slice or print, we are still telling it what to do. It is not a living creature therefore it is a tool for us to use just as is a drill, scalpel, soldering iron or pliers.

 

Cheers

Cav

[Guest Natalie Graham]

So how does CAD designing a bodyshell for a loco and having it produced ona 3D printing machine differ in principle to using the same CAD design methods to have a machine produce the moulds for injection moulding, for example? If CAD design for 3D printing produces scratchbuilt models can Hornby similarly claim their bodyshells are all scratcbuilt? How is the product of a 3D printer any more scratchbuilding than the products of a whitemetal casing machine or an injection moulding machne?

 

Surely the term scrathbuilt must imply that the modeller actually builds it, not designs it on a computer to be produced on a machine somewhere and the first contact he has with the model as a physical entity is to unpack it when it arrives in the post.

[Katier]

For me scratch building is something that cannot EASILY be replicated.

 

3D printing is bespoke designed because you can just tell the computer to print 50 of them and away it goes ( assuming the printer has a facility to kick each finished model off the plate automatically ).

 

3D printing falls into the same area as Resin Casting or whitemetal casting where a series of parts can be assembled. Now in some cases those parts may need to have some scratchbuilding adding to them ( for example a chassis ) but the parts themselves are bespoke designed parts rather than scratchbuilt. The Final model could be described as scratchbuilt if the builder used significant quantities of plain, hand fabricated, material to finish the model off.

 

 

[m0rris]

Personally I believe that 3D cad /printing is scratchbuilding, for the first one to roll off the line it takes just as much effort as to get a model built from scratch...

 

However, there is one important point which is probably that hardest to actually define, what is scratchbuilding? If you look at the blogs on here very few people actually truly "scratchbuild" but 99% of people use products readily availible - embossed plasticard, romford wheels, cast parts. Whereas, arguably, 3D printing (or at least the initial product of it) is possibly scratchbuilding in its "truest form" as plastic powder is converted into shells/shapes.

 

The second point is that of button pushing and this brings it forwards to the philosophical level of is it scratchbuilding when you can just push the print button? This would, bizarrely, indicate that the scratchbuilding is every point up to the point of printing which is, in old terms creating the master, where the skill is at. After that it is the same as pouring resin into a mold, i.e. replication of the master.

 

m0rris

[Guest Natalie Graham]

Is producing 3D Artwork comparable to producing a pattern for casting? The latter involves building a physical item wheres the former, while no doubt requiring specific skills is not building anything, it is designing the part for machine production whether that be for a run of 10, 000 or just one.

[54Strat]

Taking scratch building at face value, it's about creating a model from raw materials with no processed components. In reality this is rarely achieved. For example, turned brass buffers, extruded plasticard sections and even transfers are all processed parts that we all use, but I'd argue you're not building the model from scratch are you? In fact, it's more like a hybrid, bit's from here, bit's from there, custom parts, raw materials and whatever works is the actual reality. This is a perfectly acceptable re-definition of the term scratch building.

 

So therefore in this context, 3D printing is part of this process. We're manufacturing a component that is partly used to create a model from scratch. Manufacturing is the key word here, like the turned brass buffers we use, they are manufactured. Scratch building is itself a manufacturing process. They're the same.

 

As someone who's worked in the 3D / CAD industry for the past fifteen years, I find it quite interesting the motives for attaching a label to this process. Maybe it's less about self achievement and more about about kudos? Of course, this is assuming that scratch building is a more aspirational technique.

 

Paul

[Katier]

 

 

So therefore in this context, 3D printing is part of this process. We're manufacturing a component that is partly used to create a model from scratch. Manufacturing is the key word here, like the turned brass buffers we use, they are manufactured. Scratch building is itself a manufacturing process. They're the same.

 

 

That is assuming you are making components. Take tractions fantastic t-gauge printed class 67.. Is that scratchbuilt? Of course not it's a designed body using the same technicques that may go into a 'template' for a mould for extrusions at Hornby etc.

 

 

On the flip side if you are building parts that are just a subcomponent of a model to which you are adding a mix of parts from other sources then I would say it is scratchbuilding.. but the model.. not the 3D printed components.